For Better Factory Audits, Manage Bias (And Send Women)
All audit teams aren’t equal. Four actions can make yours more effective.
Companies can’t outsource risks to their reputation. If suppliers violate human rights or have dangerous working conditions, companies are often held responsible by activists, investors, and other stakeholders.
For example, Nike
, and other companies have faced public backlash for child labour and poor working conditions in their contractors’ overseas factories. Aware of the risk to their brand reputation, many companies have created codes of conduct and audit systems for the factories that manufacture their products.
But audits don’t always reveal the truth. A fire in a textile factory
in Karachi, Pakistan killed hundreds in 2012, partly because of locked emergency exits. Auditors had certified the factory’s safety standards about two months earlier.
Researchers Jodi Short
(University of California, Berkeley), Michael Toffel
(Harvard University) and Andrea Hugill
(Georgetown University) studied the audits of nearly 6,000 suppliers, mostly garment factories.
Their research shows that auditors are human. As they walk through a factory, observing workers, a variety of biases or social factors affect their judgments.
The result: Audit teams reported an average of 6.5 violations per factory. But the number of violations reported went up or down depending on the audit team’s experience, funder, gender makeup, and training.
The authors recommend clear steps that companies can take to help make sure more violations are reported:
- Pay for the audit: When a company paid for its supplier’s audit, auditors reported 8% more violations than when the factory paid. This finding suggests that when a factory pays for its own audit, the money biases the result, with auditors becoming more lenient.
- Send fresh faces: Audit teams reported 4% fewer violations if even a single one of their team members had audited that factory before. The authors suggest familiar auditors might become too familiar with the factories they are assessing, and that fresh perspectives can identify overlooked problems.
- Send women: Audit teams with women on them reported 7% more violations than all-male teams. In particular, women were more likely to report issues related to forced labour, working hours and wages. Female auditors may identify more violations partly because workers at garment factories tend to be women who might be more comfortable sharing information with other women. Research also shows that women work differently than men. For example, women are more likely to follow rules and they are better at understanding others’ emotions from their expressions.
- Train auditors: Auditors with more in-house training in auditing reported more violations. This was only true for specific training in auditing, not broader education.
For managers, the lesson from this study is that hiring an auditing firm and leaving the rest to them is a mistake. The authors recommend asking your auditing firm questions about how they build their auditing teams. Know how auditors are rotated so that they don’t become too familiar with a company. Find out if there are experienced, well-trained auditors on every team who can spot subtle clues that there are problems at a factory. And ask — are there women?